In its inception the internet was decentralized. Meanwhile the client-server model slowly drove us away from that premise. Today's clouds are centralized, they live on data centers usually controlled by a single entity.
We believe the centralized model is just a hindrance along the way, that arose due to difficulties of the past. Decentralized systems are more robust, and can deliver better performance. Think of BitTorrent.
We are also seeing the rise of crypto-currencies. They embrace decentralization. Nevertheless they face some hurdles on the way to mainstream adoption. One of the bigger problems is the perceived lack of intrinsic value. There is no state with its institutions nor any asset that makes a cryptocurrency its legal tender. Cryptocurrencies that keep the promise of decentralization don't have any backing in gold or any other assets except for brand recognition.
This is not just a theoretical problem. The prices of cryptocurrencies are very volatile which while good for speculation is one of the factors preventing mainstream adoption.
We believe these problems have a common solution.
By using cryptocurrency we can motivate people to lend their machines to a distributed cloud platform. In turn by building a decentralized cloud we build a tangible sought after resource that can be used as a solid yet distributed backing for the value of a cryptocurrency, provided we make its emission mechanism tied to the supply of the resources instead of new block creation.
Welcome to Nebula Network.
No, it's an entirely new blockchain.
To make Nebula work we need a (micro) transaction throughput far beyond of what the currently proven blockchain
(i.e. Ethereum and Bitcoin) solutions offer. That is why we are developing our brand new Multichain technology.
Furthermore a critical economical aspect of Nebula is its resource availability related emission mechanism. In most other cryptocurrencies all emission is tied to production of new blocks (thus mining or staking) which is not the case with Nebula.
You have a valid concern - modern data centers are carefully built by talented engineers, and it might be impossible to match their
raw power with a system composed of ad-hoc machines. Nevertheless the data centers are few and thinly spread in comparison to all other
computers around the globe. It is those geographic and topological advantages we seek to use out to our advantage.
Also note that unused machines in a data center can also join the Nebula Network.
And remember that building a truly distributed platform is about much more than just raw power. For us it's about reliability and independence. For us it's about freedom.
First of all, the hashes of the blocks within the subchains after some time get to a master chain tying it all together so that long running
attacks on single subchains are impossible.
Furthermore we have a mechanism we call collateral staking or proof of collateral that makes 51% attacks more expensive economically. When a double spend is detected we can reimburse the receiver with the stakers collateral. Used together with network size assessment algorithms and bloom filter based new block & staker gossip (to prevent staking many blocks with the same capital, and attacks on the stakers themselves) practically impossible.
If you are a blockchain, cryptography or network security expert and would like to audit our solution please contact us. We really appreciate your help, and would love for this to be a community driven project.
We want to secure seed funding and prevent early forks. Nevertheless our project will be open source, and we will try to release all IP at some time - we lean towards a stretch goal in our crowdfunding. If you have any input on that please let us know.
You can look at it in two ways:
1. We are developing a distributed supercomputer where everybody can join with their device, and get paid for doing so.
2. We are developing a new blockchain-based cryptocurrency which with time aims to be rock stable as through having an emission mechanism tied to providing tangible resources.
No - computations are performed to mine Bitcoins - they are not used for anything but securing the system via a system called proof of work. You cannot use the bitcoin network to do your computations.
Golem is a very cool project indeed, and some of our goals are similar, nevertheless:
1. Golem is not really a crypto currency, and it uses Ethereum in the background. Neither Ethereum nor Golem token emission mechanisms are in any way tied to the amount of computational power available. Nebula's main emission mechanism is tied to supply of resources such as computational power.
2. Golem seems to have a centralized application registry where you need to put your app to be whitelisted. We do not plan to control what code you can run on the system.
3. Golem seems not to have a solution for storage yet.
Our storage system is somewhat similar, nevertheless:
1. We tie providing space with the cryptocurrency emission mechanism.
2. Our organization is in no way a middleman between the service providers and users - your data is safe even from us.
If we can see further it's only because we stand on the shoulders of giants such as Bitcoin and Ethereum developers. Due to vast differences it was easier to start Nebula Network as a new codebase, nevertheless we are using various carefully selected open source libraries, and stealing various ideas from Bitcoin, Ethereum and other crytocurrencies as well as P2P projects, such as TOR and I2P. We are also using code from projects such as PNaCl so we are not developing critical pieces such as the computational sandbox from scratch.
Let us start by stating that the technology, architecture, and ideas expressed within concern us more than the language in which
we express them. That being said we want Nebula Network to be performant, and to integrate well with technologies such as LLVM and PNaCL.
Being a big project we wanted a type safe, statically typed language - as lovely as loose typing is for small projects it becomes
problematic within larger projects and requires much more vigorous and tedious testing
(which eliminated Python - which we all hold in very high regard).
We wanted a language with tight control over the resources so probably one with no automatic garbage collection, but supports
high level abstractions which meant no go for Go.
We also wanted a popular language that has a big enough coder base to be able to expand the team quickly, and good tool support on all
major platforms (IDE's & debuggers). One which we ourselves knew well. Which eliminated Rust, but we keep looking at as it seems
to be growing in popularity.
Finally we chose modern C++, that is C++ with features introduced in the 11 and 14
standard revisions. We are aware of the language shortcomings
nevertheless with the aforementioned standard revisions, C++ is also a surprisingly expressive and modern language.
Code being executed on Nebula Network on the other hand can be written in any language that compiles to LLVM, and we plan to provide SDK's and tutorials for many languages.
All the code will be available on github at the time of our first alpha release.
We're setting up a forum where we will be able to discuss all such matters. Meanwhile please dont hesitate to contact us.
We are using the sandbox form Google's PNaCl project.
In a way Nebula Network can be thought of as a self administrating grid. Each resource providing node in the network performs just a chunk of the work, alas redundantly to provide adequate security. It is worth noting that within Nebula Network the nodes are controlled by different people or organizations and they do not have to trust each other, nor do they need to trust the user, who also does not have to trust the nodes. Also nodes within the Nebula Network get a reward for providing their resources.
Nebula Network is more than just another cryptocurrency. What makes it unique, is solid backing in cloud resources via a mechanism of emission tied to their supply. Those give Nebula real value and enforces the flow of money to those who provide valuable assets.
One of the main goals of Nebula Network is to be economically stable.
To ensure the economical models we use are rock solid and we engaged dr. Moszoro.
He held academic positions at the Haas School of Business, University of California,
Berkeley, Harvard Law School, Université Paris-Dauphine, George Mason University, and Cornell University.
In 2005-2006, he was appointed Undersecretary of State and Deputy Minister of Finance of Poland,
the youngest ever in that position.
His research encompasses project finance, risk perception, and asset pricing. Having a passion for cutting-edge research and teaching, students voted his MBA course the best multiple times.
Machine Learning Advisor
We are undergoing an AI revolution right now with deep learing at the forefront. Trainig AI models can take a lot of power, and we want to optimize Nebula Network to be a great platform for that. Prof. Cichocki currently the Head of Laboratory for Advanced Brain Signal Processingat RIKEN Brain Science Institute in Japan is a world-leading scientist in the area of artificial intelligence and biomedical applications of advanced data analytics technologies with publications cited over 31000 times. He is also a professor at Center for Computational and Data-Intensive Science and Engineering at Skoltech.
Network Systems Advisor
Nebula Network is a network after all and to make sure it works well we brought in Łukasz. Łukasz is a world class network technology expert currently serving as the the CTO of Cisco Systems Poland. Holding a double CCIE and CCDE he has over a decade of telecommunication solutions building experience. He is also the co-creator of the PLNOG conference and the leader of the BGP Blackholding PL.
Network Systems Advisor
Testing and optimizing distributed systems is not an easy task and often done with the help on many virtual machines. As experienced architect Piotr gained his experience working for companies all accross the telecomunication solutions spectrum including CISCO and (currenlty) VMWare, Piotr is a perfect advisor to help us with that. Piotr holds a CCDE, CCIE, and VCP-NV, and has many years of hands-on experience with virtualization and network technology.
Like our ideas? Disagree?
Want to join our team or invest?